Optimization Method for Shipbuilding Product Manufacturing Process Based on Big Language Model
XU Shengchao
LYU Junmin
JIANG Darui
LI Xiangyuan
MAO Mingyang
JIANG Jinling
GUO Tao
Abstract：An artificial intelligence big language models based shipbuilding product manufacturing process optimization ap-proach is proposed in this paper.Inputting the design requirements of the ship into the artificial intelligence big model,the artificial intelligence big model plans the ship modules based on design requirements,module functions,ship layout,design specifications,etc.to obtain the modular design diagram of the ship,and then carries out modular manufacturing to effectively achieve multi-disci-plinary collaborative management.In the modular manufacturing process,YOLO neural network is used to detect the weld seams connected to ship modules.The weld seam image is used as input to extract weld seam features and detect the quality of the weld seams,achieving quality management in the ship manufacturing process.The experimental results show that ships designed with ar-tificial intelligence large models can travel about 0.3 kilometers more per ton of fuel.And the average coefficient of variation in the production process is as high as 80%.YOLO8 is used to inspect welds can accurately identify defects and effectively identify differ-ent weld defects.
Keywords：manufacturing processartificial intelligencelarge modelship designquality managementneural network
Publication Date：2025-09-20
Online Publishing Date：2025-12-23(First online date of this platform, not the publication date of the document)
Pages：6( 57-61,105 )